Compression. Squished, and then squished a little more!

We have all heard the hyperbole. Analog forever. The pure sound of analog.. Need I keep going. Yes all of you in the front row are nodding and smiling. The ones in the back are wondering. What is this guy up to???

I'm not for one second saying that you cannot get great sound from a record. Nor am I slyly trying to say that digital recordings are the be all and the end all.

I'm interested in the means and the methods used in service of either and or all of the above.

After all I enjoy listening to music. I'm guessing that you do to.



​

---

---

Let's get a little closer:

---

Now, really really close!

---

Those three pictures of canyons are record grooves. The first and second include the stylus and the third picture is magnified to give you a sense of the groove itself. Number three also shows a fair bit of wear. Something that is unavoidable in the use and listening of phonograph recordings.

So I started off this little blurb with a statement about compression.
​What do I mean? 

Well lets start with how a recording is prepared for being cut onto record master on a record cutting lathe. You may think that microphone, perhaps pre-amplifier to power the mics, and then direct to a cutting lathe. Remember the direct to disc recordings that started in the later 70's? Well there was a circuit in between the microphone pre-amplifiers and the cutting lathe. So what does this circuit do?

Behold the lines:

---

---

So to make a vinyl recording an equalization curve of the shape you see in the dashed blue line is created. A very simplified explanation is that the left side is the bass the right side is the treble. The line at 1000 hertz is within the beginning of our ears start most sensitive range. The vertical range is interesting. At 1000 hertz there is no equalization. At 20 hertz, fairly low in frequency there is a reduction of minus 20 decibels. It usually takes a 10 decibel cut in volume when listening to a speaker to say it is half as loud. So roughly a 20 decibel cut is going to sound one fourth as loud. To the right side of the graph you will see a boost of 40 decibels in relation to the cut at 20 hertz. That is a ratio of 10000 to 1 remembering that decibels are logarithmic. Not a linear scale as you would see on a ruler, or a tape measure . So the amount of boost and cut is not trivial. There is some serious compression of the signal.

Now this leads me to the mechanical side of a phonograph recording. Those pictures at the beginning of this post. Imagine how this works. Side to side and vertical movement of a single point cutter on a lathe will be used to convey the recording. As simple as it is complex. Record lathes are very complex pieces of equipment. The good ones are massive and cost near the same as a decent house. 

So lets recap. Our input signal from the microphones or master tape is really compressed on the low end and really boosted on the high end of the recording spectrum. Sent to a very highly engineered cutting lathe that creates a record master. (I'll leave out the negative creation and all that stuff) Then to play it back you take a needle that follows a single groove vibrating to the left and the right and a little bit vertically. That signal gets equalized again this time in the fashion of the red curve. 40 decibels of boost at 20 hertz and 20 decibels of cut at 20000 hertz.

And you listen to it and marvel at how great it sounds. Some of the highest mechanical and electrical compression that mankind has been able to repeatedly manufacture. Record needles that may or may not be flat in their response. Playback equalization that again should follow a prescribed RIAA playback curve but not always. A mechanical playback method that by it's very design is susceptible to extraneous vibration. And a storage medium that from the first time you play the record you are wearing out a little at a time.

Yep the analog sound. It amazes me that records can sound as good as they do. But make no mistake records so have a sound. the top end generally rolls off above 15 kilohertz. Not that big a deal as there is very little musical information up that high in the first place. The low end is usually rolled off and summed to mono below a prescribed midbass frequency that can vary from studio to studio. Even from recording to recording. But there is something that has not been mentioned yet. Modulation of the sound of one groove to the sound of the  one beside it. That's right someone is humming along with your tunes ( Glen Gould need not apply). So yes there is an analog sound. And it can be rather convincing. I have listened to a displayed very high end record systems at a few RMAF shows in the past.

I'll let you in on a little secret. I can be rather mischievous at times. I have done a few experiments of my own. I'll preable this story with a little explanation of another compression format. MP3 is a lossy compression format that if you use the higher bit rate of 320 kilobits per second can be rather useful to listen to music in a bit noisier environment like an automobile. So I have about 60GB of MP3 recordings that I have made myself from my collections of compact discs. I bring them with me pretty much everywhere. So back to Colorado and the Audio Festival. I had a group of people come into my demo room and I asked them if they wanted to listen to a recording. I played with the record player and sat down at my laptop and let them listen to a recording. They waxed eloquently of how there is nothing like the purity of the sound of vinyl. They could tell right of the bat that it was a truly quality sound system. 

They left. I asked our door man to lock the door and burst out laughing. He knew there was something fishy. They listened to an MP3. I motved towards the record player. Asked if they wanted to hear a recording and lowered the needle close to the record. Then I pressed play on Foobar2000 of a previously cued MP3.

So what did I prove?

I substituted a highly compressed recording in the place of another highly compressed recording. Both had high quality sources. 

Did I prove the old saying that people listen with their eyes? Probably.
Did I prove the superiority of one recording compression method over the other? Hardly.

I simply proved that a well done recording be it made in a number of different ways can give us a rather convincing facsimile of a live music event. 

​And that my friends is what I'm after in the first place. Recreating as accurately as possible the original recording.